Many transition metal complexes have hitherto been used as catalysts for organic synthesis reactions. In particular, noble-metal complexes are extensively utilized, despite their expensiveness, since they are stable and easy to handle. Many investigations were made on syntheses using transition metal complexes including such noble-metal complexes as catalysts. As a result, many reports have been made on techniques making it possible to carry out organic synthesis reactions, including asymmetric reactions, which have been regarded as impossible with any conventional technique.
There are various types of optically active ligands for use in such asymmetric-synthesis catalysts. Among the ligands for use in asymmetric hydroformylation reactions using transition metal-phosphine complexes, one of the ligands having the highest degree of chiral recognition is 2-diphenylphosphino-1,1'-binaphthalen-2'-yloxy(1,1'-binaphthalene-2,2'-diy loxy)phosphine (hereinafter referred to simply as "BINAPHOS"). There are reports on the use of a rhodium complex containing BINAPHOS as a ligand in an olefin hydroformylation reaction, which is a reaction for forming asymmetric carbon-carbon bonds (see JP-A-6-263776 and JP-A-6-316560). (The term "JP-A" as used herein means an "unexamined published Japanese patent application".)
However, such expensive catalysts are unable to be recovered, or can be recovered only by a complicated separation method which is always accompanied by an undesirable loss. Furthermore, reuse of the recovered homogeneous catalysts is impossible and/or uneconomical. There has hence been a desire for a catalyst which can be easily recovered and reused and is capable of fully retaining its activity and, in particular, selectivity during repeated use.
With respect to synthetic chiral polymers, the application thereof to racemate separation media, reagents for asymmetric syntheses, catalysts, and the like is being extensively investigated. Rapid progress is being made recently in investigations on asymmetry recognition among the various functions of these chiral polymers. In particular, in the application thereof to stereoselective organic reactions, the chiral polymers can be used in a method different from those for general homogenous reaction systems because a specific reaction field constituted of the polymers is used.
Use of a polymeric reagent or polymeric catalyst in organic syntheses has an advantage that industrial processes can be improved because the reaction products can be easily separated and the reagent or catalyst can be reused.
For example, a report has been made on a process comprising reacting an optically active amino acid with 4-vinylbenzenesulfonyl chloride to obtain a chiral monomer, polymerizing the monomer with styrene and divinylbenzene to obtain a chiral polymer, ##STR2## reacting this polymeric ligand with diborane to obtain a polymer-supported chiral oxaborolidinone, and using this compound as a Lewis acid catalyst to conduct the Diels-Alder reaction of cyclopentadiene with methacrolein (see S. Itsuno et al., Tetrahedron: Asymmetry, 1995, Vol. 6, p. 2547).
There also is a report on a method in which a Mn(II)-salen complex is polymerized ##STR3## and the resultant polymer is used to conduct the asymmetric epoxidation reaction of an olefin (see S. Sivaram et al., Tetrahedron: Asymmetry, 1995, Vol. 6, p. 2105).
Furthermore, there is a report on a method which comprises copolymerizing optically active 2-p-styryl-4,5-bis[(dibenzophosphoryl)methyl]-1,3-dioxolane with styrene to obtain a chiral polymer, ##STR4## coordinating platinum chloride to this polymeric ligand, and using the resultant coordination compound to conduct the hydroformylation reaction of styrene in the presence of tin chloride (see J. K. Stille et al., J. Org. Chem., 1986, Vol. 51, p. 4189).
However, all the prior art techniques have insufficient catalytic activity or result in an enantiomer excess lower than those in the case of reacting monomers. Any of those prior art techniques has not been put to industrial use.